Step-by-Step Tutorial: Create a RAG Chatbot with n8n AI Agents in MINUTES

Leon van Zyl writes:

In this n8n AI Agent tutorial you will learn how to create a RAG Chatbot that can answer questions from your own knowledge base.

We will also embed the chatbot into a website using the n8n web embed script.

13 Likes

Excellent work, Leon. Do you have any tutorial on better managing knowledge data? For example, how to ensure that when a document in Drive is modified, the vector database is automatically updated? Or when a document is deleted, it is also removed from the vector database?

Best regards!

Thank you very much. Im newbie and it so helpful with me <3

Thank you for this, but I built according to instructions, and it is not working:

  • The google drive loaded 34 records, and will not load any more
  • when querying the chat about the docs, it just spins and never stops.

Thank you for this! My first successful (sorta) RAG!

Appreciate if you could help with some questions and issues I have. I did everything as per the video but still got strange responses.

I put in some company SOPs in a table format in Google Docs, it failed to read the data.
Then I made them into paragraph format, and it manages to read and return data.
After a while, it makes its own answers.. For example, first time I asked it how to prepare a Costing Sheet, it churned out a good response as per our SOP. Second time I asked, it returned a generic ChatGPT answer that deviated from our SOP.

Not sure why this happened.. Then when I put public URL, the answers became totally deviated from the SOP and it answered as generic ChatGPT…

Pretty stuck at this point cos I didn’t get any error messages.